Markov chain Monte Carlo methods for state-space models with point process observations

Yuan, K. , Girolami, M. and Niranjan, M. (2012) Markov chain Monte Carlo methods for state-space models with point process observations. Neural Computation, 24(6), pp. 1462-1486. (doi: 10.1162/NECO_a_00281) (PMID:22364499)

129185.pdf - Published Version



This letter considers how a number of modern Markov chain Monte Carlo (MCMC) methods can be applied for parameter estimation and inference in state-space models with point process observations. We quantified the efficiencies of these MCMC methods on synthetic data, and our results suggest that the Reimannian manifold Hamiltonian Monte Carlo method offers the best performance. We further compared such a method with a previously tested variational Bayes method on two experimental data sets. Results indicate similar performance on the large data sets and superior performance on small ones. The work offers an extensive suite of MCMC algorithms evaluated on an important class of models for physiological signal analysis.

Item Type:Articles (Letter)
Additional Information:K.Y. is supported by a studentship from the University of Southampton. M.G. acknowledges EPSRC Advanced Fellowship EP/E052029/2, project grants EP/E032745/2 and EP/F009429/2, and the BBSRC project grant: The Silicon Trypanasome.
Glasgow Author(s) Enlighten ID:Yuan, Dr Ke and Girolami, Prof Mark
Authors: Yuan, K., Girolami, M., and Niranjan, M.
College/School:College of Science and Engineering > School of Computing Science
Journal Name:Neural Computation
Publisher:MIT Press
ISSN (Online):1530-888X
Published Online:25 April 2012
Copyright Holders:Copyright © 2012 Massachusetts Institute of Technology
First Published:First published in Neural Computation 24(6):1462-1486
Publisher Policy:Reproduced in accordance with the publisher copyright policy

University Staff: Request a correction | Enlighten Editors: Update this record